The present invention is directed generally to an in-band fault detection and isolation system and method for networks such as Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH) networks.
In a Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH) network, signal fail (SF) and signal degrade (SD) faults are detected at a SONET/SDH network element (NE) that is nearest to the fault location. Such a fault may be caused by a loss of signal (LOS), a loss of frame (LOF), or an excessive bit error rate (E-BER) in the link. According to SONET/SDH standards, the nearest pair of SONET/SDH NEs inserts an Alarm Indication Signal (AIS-P) in the failed path upon detection of a fault so that all other NEs down the signal paths are informed that there is a fault in the upstream.
Referring to FIG. 1, in a SONET/SDH network, data and overhead is transported in the format of a synchronous transport signal (STS) frame 100. Each STS frame 100 includes a synchronous payload envelope (SPE) 101 and transport overhead bytes 102. The transport overhead bytes 102 include pointers, such as pointers H1 and H2. The SPE 101 includes path overhead (POH) bytes 103, including:
Class A bytes J1 (trace byte), B3 (bit interleaving parityxe2x80x948 byte), C2 (signal label byte), and G1 (path status byte);
Class B byte H4 (indicator byte);
Class C byte F2 (user channel byte); and
Class D bytes Z3 (reserved for growth and DQDB mapping) and Z4 (reserved for growth).
Additional bytes are included in the transport overhead bytes 102 and in the POH bytes 103, but these additional bytes need not be discussed herein.
In the conventional AIS-P, all of the bits in the bytes of the SPE 101 are set to one. As a result, in a conventional AIS-P, POH bytes 103 (which are a part of the SPE 101) do not carry valid information. Also, in a conventional AIS-P, both pointers H1, H2 are set to all ones (i.e., all of the bits in the pointer bytes H1 and H2 are set to one).
Referring to FIG. 2, a network 200 may include one or more SONET/SDH network elements (NEs). SONET/SDH NEs may include add/drop multiplexers (ADMs) (e.g., ADMs B, D, H), digital cross connects (DCSs) (e.g., DCSs A, K), and/or optical transport systems (OTSs) (e.g., OTSs C, E, F, G, 3). Each of the SONET/SDH NEs may be interconnected with one or more of the other SONET/SDH NEs via one or more links (e.g., links X-J, J-B, B-C, C-G, G-H, H-V, B-A, H-K, A-D, Y-F, F-D, D-E, E-Z). The network 200 may further include one or more path terminating equipment (PTE) M and/or network management system (NMS) N. The NMS M may be, for example, a computer. Although the PTE M is shown as being a separate piece of equipment, a SONET/SDH DCS can also be a PTE for some SONET/SDH signals if the SONET/SDH path terminates at the SONET/SDH DCS. FIG. 2 also illustrates SONET/SDH NEs X, Y, and Z, which may be, for example, ADMs and/or DCSs.
For purposes of illustrating how a conventional SONET/SDH network handles a conventional AIS-P, assume that a fault occurs in link A-B. In response to the fault, DCS A inserts an AIS-P on the bearer tributaries going towards ADM D (i.e., via link A-D). ADM D receives the AIS-P on its appropriate OC-N ports. Recognizing the AIS-P, ADM D passes the AIS-P through on the connected port(s) to the downstream signal path. Each ADM and DCS downstream on the signal path from ADM D also passes the AIS-P through. These downstream DCSs and ADMs do not raise alarms for the paths.
Also, on the other side of the fault in link A-B, ADM B inserts an AIS-P in the signal on link B-C towards ADM H and/or on the link B-J towards network element X. The AIS-P is received by ADM H and passed through by ADM H and then by DCS K, eventually to a PTE, which assembles the SONET/SDH path.
Thus, the PTE receives the AIS-P. In response to receiving the AIS-P, the PTE inserts the AIS into the signals on the downstream constituent link (e.g., a DS3 link, DS1 link, etc.) and may also send a SONET/SDH path failure alarm to the network management system NMS if provisioned to do so.
As a result, all NEs in the network 200 receiving the AIS-P recognize that there is a fault somewhere in the signal path. However, none of the NEs, except the pair nearest to the fault (in the above scenario, ADM B and DCA A),: can identify where the fault has occurred. Moreover, NEs do not retain any information about the location of the fault. Fault isolation is left to the NMS receiving the alarms from all NEs under its domain.
Furthermore, in some network configurations, particularly those without rings, NEs such as ADMs adjacent to the fault location are not capable of restoring the failed or degraded signal through alternate routes. Generally the task is carried out by DCSs connected to ADMs via facility-level SONET/SDH signals. If a DCS is not adjacent to the fault location, however, it must be informed by the NMS about the fault location. This method of fault isolation and dissemination of the information to the NEs for the purpose of signal restoration is time consuming and expensive.
The above-described problems are solved by implementing various aspects of the present invention. According to one aspect of the present invention, fault detection and isolation may be quickly and efficiently performed by a network without requiring modification to any add/drop multiplexers (ADMs). This is practical because ADMs already exist in many current networks. Furthermore, fault detection may be quickly and efficiently performed without the involvement of an external network management system (NMS).
According to another aspect of the present invention, the present invention may be implemented without violating any existing Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH) standards. To accomplish this, a new signal, called a fault isolation signal in the path (FIS-P), may be used. A FIS-P may be used in lieu of an AIS-P to support special applications on certain digital cross connect (DCS)-to-DCS path segments. A FIS-P may include an unused SONET/SDH path overhead (POH) byte that may be utilized for quickly isolating faults without the use of external monitoring systems. This byte, called a xe2x80x9cfault isolation byte,xe2x80x9d may be any byte in the synchronous payload envelope (SPE). However, the Z4 byte (which, as discussed above, is one of the POH bytes) is an optimal choice for a fault isolation byte because the Z4 byte is completely unused by SONET/SDH for any other purpose.
According to yet another aspect of the present invention, a network may implement the present invention without changing the behavior of the path terminating equipment (PTE). This may be accomplished by implementing one or more of several variations of the present invention, as presented below. According to this aspect of the present invention, three examples of embodiments of such variations of the invention are presented below, including the xe2x80x9cTime-Outxe2x80x9d embodiments, the xe2x80x9cFIS-P Terminationxe2x80x9d embodiments, and the xe2x80x9cFIS-P/AIS-P Flipxe2x80x9d embodiments.
Further aspects of the invention may be directed to: a method for detecting a fault in a SONET/SDH network, the method comprising the steps of receiving in a first network element a fault indication in the form of at least one of a failed signal, a degraded signal, and an AIS-P; and generating in the first network element an FIS-P in response to the step of receiving the fault indication, the FIS-P. comprising an SPE having a plurality of bytes including a plurality of POH bytes, each byte comprising a plurality of bits, wherein all of the bits in the bytes of the SPE except for at least one of the plurality of POH bytes are set to all ones.
Further aspects of the invention may be directed to: in a SONET/SDH network, a method comprising the steps of receiving by a downstream network element a signal from an upstream direction; and generating by the downstream network element a complementary FIS-P in the upstream direction responsive to receiving the signal.
Further aspects of the invention may be directed to: a network element configured to be connected in a SONET/SDH network and further. configured to generate a fault isolation signal in the path (FIS-P) in response to receiving at least one of a failed signal, a degraded signal, and an AIS-P.
Further aspects of the invention may be directed to. a network element configured to be connected in a SONET/SDH network and further configured to generate a complementary FIS-P in an upstream direction in response to receiving a signal from the upstream direction.
Further aspects of the invention may be directed to: a network element configured to be connected in a SONET/SDH network and further configured to generate an AIS-P in a downstream direction in response to receiving a complementary FIS-P from the downstream direction.
Further aspects of the invention may be directed to: a method for detecting a fault in a SONET/SDH network, the method comprising the steps of receiving in a network element a fault indication in the form of at least one of a failed signal, a degraded signal, and an AIS-P; and generating in the network element an FIS-P in response to the step of receiving the fault indication, the FIS-P being configured such that a SONET/SDH ADM receiving the FIS-P will pass the FIS-P through without detecting any change in signal status that is due to an existence of the FIS-P and a SONET/SDH DCS receiving the FIS-P will recognize an existence of a fault in the SONET/SDH network by reading the FIS-P.
Further aspects of the invention may be directed to: a digital cross connect configured to be connected in a SONET/SDH path, the SONET/SDH path having path termination equipment (PTE), the digital cross connect being configured to pass through an FIS-P in response to receiving the FIS-P if the network element is not a last DCS on the SONET/SDH path nearest to the PTE; and generate an AIS-P in response to receiving the FIS-P if the network element is the last DCS on the SONET/SDH path nearest to the PTE.